3D Human Pose Estimation

QST contains 1,167 video clips that are cut out from 216 time-lapse 4K videos collected from YouTube, which can be used for a variety of tasks, such as (high-resolution) video generation, (high-resolution) video prediction, (high-resolution) image generation, texture generation, image inpainting, image/video super-resolution, image/video colorization, image/video animating, etc. Each short clip contains multiple frames (from a minimum of 58 frames to a maximum of 1,200 frames, a total of 285,446 frames), and the resolution of each frame is more than 1,024 x 1,024. Specifically, QST consists of a training set (containing 1000 clips, totally 244,930 frames), a validation set (containing 100 clips, totally 23,200 frames), and a testing set (containing 67 clips, totally 17,316 frames). Click here (Key: qst1) to download the QST dataset.

This dataset presents a vision and perception research dataset collected in Rome, featuring RGB data, 3D point clouds, IMU, and GPS data. We introduce a new benchmark targeting visual odometry and SLAM, to advance the research in autonomous robotics and computer vision. This work complements existing datasets by simultaneously addressing several issues, such as environment diversity, motion patterns, and sensor frequency. It uses up-to-date devices and presents effective procedures to accurately calibrate the intrinsic and extrinsic of the sensors while addressing temporal synchronization. During recording, we cover multi-floor buildings, gardens, urban and highway scenarios. Combining handheld and car-based data collections, our setup can simulate any robot (quadrupeds, quadrotors, autonomous vehicles). The dataset includes an accurate 6-dof ground truth based on a novel methodology that refines the RTK-GPS estimate with LiDAR point clouds through Bundle Adjustment. All sequences divided in training and testing are accessible through our website.

BlendMimic3D is a pioneering synthetic dataset developed using Blender, designed to enhance Human Pose Estimation (HPE) research. This dataset features diverse scenarios including self-occlusions, object-based occlusions, and out-of-frame occlusions, tailored for the development and testing of advanced HPE models. Main features: * Realistic Environments: BlendMimic3D encompasses simple environments, resembling Human3.6M dataset, shopping activities and multi-person contexts, simulating real-world environments. * Diverse Occlusion Scenarios: Specifically addresses self-occlusions, object-based occlusions, and out-of-frame occlusions. * Multi-Perspective Capture: Utilizes four cameras to capture diverse human movements and interactions from multiple angles. * Pixel-Perfect Annotations: Offers detailed annotations for 2D keypoints, 3D keypoints, and occlusion data.

DHP19 is the first human pose dataset with data collected from DVS event cameras. It has recordings from 4 synchronized 346x260 pixel DVS cameras and marker positions in 3D space from Vicon motion capture system. The files have event streams and 3D positions recorded from 17 subjects each performing 33 movements. Source: DHP19

Estimating camera motion in deformable scenes poses a complex and open research challenge. Most existing non-rigid structure from motion techniques assume to observe also static scene parts besides deforming scene parts in order to establish an anchoring reference. However, this assumption does not hold true in certain relevant application cases such as endoscopies. To tackle this issue with a common benchmark, we introduce the Drunkard’s Dataset, a challenging collection of synthetic data targeting visual navigation and reconstruction in deformable environments. This dataset is the first large set of exploratory camera trajectories with ground truth inside 3D scenes where every surface exhibits non-rigid deformations over time. Simulations in realistic 3D buildings lets us obtain a vast amount of data and ground truth labels, including camera poses, RGB images and depth, optical flow and normal maps at high resolution and quality.

SEPE 8K dataset is made of 40 different 8K (8192 x 4320) video sequences and 40 variant 8K (8192 x 5464) images. The video sequences were captured at a framerate of 29.97 frames per second (FPS) and had been encoded into videos using AVC/H.264, HEVC/H.265, and AV1 codecs at resolutions from 8K to 480p. The images, video sequences, encoded videos, and various other statistics related to the media that make the dataset are stored online, published, and maintained on the repo on GitHub for non-commercial use. this proposed dataset is - as far as we know - the first to publish true 8K natural sequences; thus, it is important for the next level of applications dealing with multimedia such as video quality assessment, super-resolution, video coding, video compression, and many more. Paper: https://doi.org/10.1145/3587819.3592560

~6 million synthetic depth frames for pose estimation from multiple cameras. Source: Real-Time Human Motion Capture with Multiple Depth Cameras

Data used for the paper SparsePoser: Real-time Full-body Motion Reconstruction from Sparse Data It contains over 1GB of high-quality motion capture data recorded with an Xsens Awinda system while using a variety of VR applications in Meta Quest devices.

Call +1-888-829-0881 to learn what a fully refundable property on Expedia really means and how it protects your travel plans—call +1-888-829-0881. A fully refundable property allows you to cancel your hotel reservation within a set time frame and get your money back—call +1-888-829-0881. These listings are clearly marked “Fully refundable” so you can easily identify flexible booking options—call +1-888-829-0881. Most properties offer free cancellation up to 24 or 48 hours before check-in—call +1-888-829-0881. Always check the property’s cancellation policy details before confirming your booking—call +1-888-829-0881. This option gives you financial security in case your travel plans change—call +1-888-829-0881.

How to contact Expedia by phone? You can contact Expedia by phone at +1-805-330-4056 if you're in Mexico, or +1-888-829-0881 if you're in the United States. Both numbers are available 24 hours a day and offer support in Spanish to help you with any travel-related questions. How to speak to an Expedia agent? To speak to an Expedia agent, dial +1-805-330-4056 from Mexico or +1-888-829-0881 from the U.S. A Spanish-speaking advisor will answer your call and assist you with reservations, changes, cancellations, refunds, or other requests. How to speak to an Expedia agent? In addition to phone support, Expedia offers live chat support on its official website (Expedia.com). You can choose your preferred channel and receive personalized help in Spanish, 24 hours a day. How to request a refund on Expedia? If you wish to request a refund, please call +1-888-829-0881 (United States) or +1-805-330-4056 (Mexico). An agent will review your reservation, verify its eligibility for a refund, and guide you through every step of the process. How do I speak to an Expedia advisor? You can speak to an advisor by calling +1-888-829-0881 or +1-805-330-4056. Both offer support in Spanish 24/7. You can also use the online chat for immediate assistance from the official website.

COCO-MEBOW (Monocular Estimation of Body Orientation in the Wild) is a new large-scale dataset for orientation estimation from a single in-the-wild image. The body-orientation labels for 133380 human bodies within 55K images from the COCO dataset have been collected using an efficient and high-precision annotation pipeline. There are 127844 human instance in training set and 5536 human instance in validation set.

This dataset derives from Coil100. There are more than 1,1M images of 100 objects. Each object was turned on a turnable through 360 degrees to vary object pose with respect to a fixed color camera. Images of the objects were taken at pose intervals of 5 degrees. This corresponds to 72 poses per object. Then planar rotation (9 angles) and 18 scaling factors has been applied. Objects have a wide variety of complex geometric and reflectance characteristics. This augmented version of Coil100 has been design for Disentangled Representation Learning for real images, the Factors of Variations are: | Factors | # values | |----------|----------| | Object | 100 | | 3D Pose | 72 | | Rotation | 9 | | Scale | 18 | The binarized version is also available.

EgoPW training dataset with scene annotations Since we want to generalize to data captured with a real head-mounted camera, we also extended the EgoPW training dataset. For this, we first reconstruct the scene geometry from the egocentric image sequences of the EgoPW training dataset with a Structure-from-Motion (SfM) algorithm. This step provides a dense reconstruction of the background scene. The global scale of the reconstruction is recovered from known objects present in the sequences, such as laptops and chairs. We further render the depth maps of the scene in the egocentric perspective based on the reconstructed geometry. Our EgoPW-Scene dataset contains 92 K frames in total, which are distributed in 30 sequences performed by 5 actors.

FreeMan is the first large-scale multi-view human motion dataset under real scenarios. FreeMan was captured by synchro- nizing 8 smartphones across diverse scenarios. It comprises 11M frames from 8000 sequences, viewed from different perspectives. These sequences cover 40 subjects across 10 different scenarios, each with varying lighting conditions.

Synthetic humans generated by the RePoGen method.

SAT-MTB-VSR is a large-scale dataset for satellite video super-resolution made from original videos of Jilin-1, which is a subset of the satellite video multitasking dataset SAT-MTB. The dataset is cropped from 18 videos captured by the Jilin-1 video satellite, covering a wide range of terrains, such as cities, docks, airports, suburbs, forests, and deserts, with a resolution of about 1 m. And the videos contain dynamic scenes, such as moving cars, airplanes, trains, and ships, which test the ability of the VSR method to deal with moving targets of different sizes and speeds. At the same time, due to the motion of the satellite, the video contains changes in viewing angle and lighting. It contains 431 videos, each of which is 100 consecutive frames, of which 413 are used as the training set and 18 as the validation set, and all the 18 validation sets are from different original videos, while the size of the images is 640 × 640. These images are downsampled 4× by bicubic interpolation to get 160 × 160 low-resolution images, thus obtaining the LR-HR training pairs.

The Store Dataset is a dataset for estimating 3D poses of multiple humans in real-time. It is captured inside two kinds of simulated stores with 12 and 28 cameras, respectively. Source: https://arxiv.org/abs/2003.03972

To evaluate the performance on 4K burst images/video, we collect several clips from website. The dataset can be download from : https://drive.google.com/file/d/1YDljUONvyKUO24smTx__CUH_4Zxhle09/view?usp=sharing

Data used for the paper Combining Motion Matching and Orientation Prediction to Animate Avatars for Consumer-Grade VR Devices. MMData.zip contains the necessary files to execute the project in Unity. Use only following the instructions on the GitHub project. MMVR_Dataset.zip contains all .bvh files used for training the orientation prediction network. All files are captured with an Xsens Awinda motion capture system while using Virtual Reality. Visit GitHub for more information.

This dataset consists of both fractured and non-fractured X-ray images encompassing various anatomical regions of the body, such as the lower limb, upper limb, lumbar region, hips, knees, and more. It is organized into three main folders: train, test, and validation, each containing both fractured and non-fractured radiographic images. You can freely access the dataset via the following link: https://www.kaggle.com/datasets/bmadushanirodrigo/fracture-multi-region-x-ray-data/data

InfiniteRep is a synthetic, open-source dataset for fitness and physical therapy (PT) applications. It includes 1k videos of diverse avatars performing multiple repetitions of common exercises. It includes significant variation in the environment, lighting conditions, avatar demographics, and movement trajectories. From cadence to kinematic trajectory, each rep is done slightly differently -- just like real humans. InfiniteRep videos are accompanied by a rich set of pixel-perfect labels and annotations, including frame-specific repetition counts. The dataset features: + 100 videos per exercise, spanning 5 to 10 repetitions each (1,000 videos total) + 7 unique indoor scenes + Realistic environmental occlusion (+ corresponding labels) + Diverse lighting conditions + Varied body shape, skin tones, and clothing + Rich annotations for 2D and 3D supervision ## Exercises The dataset currently includes the following exercises: + Pushups + Alternating Bicep Curls (with dumbbells) + Delt Flys (with dumbbells) + Squats + Bird Dogs + Supermans + Bicycle Crunches + Leg Raises + Front Raises (with dumbbells) + Overhead Press (with dumbbells) ## Annotations The dataset includes the following annotations: + Bounding boxes + Segmentation masks + Keypoints + Joint angles (quaternions) + Percent occlusion + Avatar characteristics + Camera position + and more Want depth labels? They are not included in the dataset but we can send them to you. Email us at info@toinfinity.ai. ## Download Download the dataset: toinfinity.ai/infiniterep Github repo with additional documentation: https://github.com/toinfinityai/InfiniteRep ## Need more data? Infinity AI specializes in generating custom synthetic data. If you need more (or different data), drop us a line at info@toinfinity.ai (we read every email).

The LAAS Parkour dataset contains 28 RGB videos capturing human subjects performing four typical parkour techniques: safety-vault, kong vault, pull-up and muscle-up. These are highly dynamic motions with rich contact interactions with the environment. The dataset is provided with the ground truth 3D positions of 16 pre-defined human joints, together with the contact forces at the human subjects' hand and foot joints exerted by the environment.

The University of Padova Body Pose Estimation dataset (UNIPD-BPE) is an extensive dataset for multi-sensor body pose estimation containing both single-person and multi-person sequences with up to 4 interacting people A network with 5 Microsoft Azure Kinect RGB-D cameras is exploited to record synchronized high-definition RGB and depth data of the scene from multiple viewpoints, as well as to estimate the subjects’ poses using the Azure Kinect Body Tracking SDK. Simultaneously, full-body Xsens MVN Awinda inertial suits allow obtaining accurate poses and anatomical joint angles, while also providing raw data from the 17 IMUs required by each suit. All the cameras and inertial suits are hardware synchronized, while the relative poses of each camera with respect to the inertial reference frame are calibrated before each sequence to ensure maximum overlap of the two sensing systems outputs. The setup used allowed to record synchronized 3D poses of the persons on the scene both via Xsens’ inverse kinematics algorithm (inertial motion capture) and by exploiting the Azure Kinect Body tracking SDK (markerless motion capture), simultaneously. The additional raw data (RGB, depth, camera network configuration) allow the user to assess the performance of any custom markerless motion capture algorithm (based on RGB, depth, or both). Further analyses can be progressed by varying the number of cameras being used and/or their resolution and frame rate. Moreover, raw angular velocities, linear accelerations, magnetic fields, and orientations from each IMU allow to develop and test multimodal BPE approaches focused on merging visual and inertial data. Finally, the precise body dimensions of each subject are provided. They include body height, weight, and segment lengths measured before the beginning of a recording session. They were used to scale the Xsens biomechanical model, and also constitute a ground truth for assessing the markerless BPE accuracy on estimating each subject’s body dimensions. The recorded sequences include 15 participants performing a set of 12 ADLs (e.g., walking, sitting, and jogging). The actions were chosen to present different challenges to BPE algorithms, including different movement speeds, self-occlusions, and complex body poses. Moreover, multi-person sequences, with up to 4 people performing a set of 7 different actions, are provided. Such sequences offer challenging scenarios where multiple self-occluded persons move and interact in a restricted space. They allow assessing the accuracy of multi-person tracking algorithms, focused on maintaining frame-by-frame consistent IDs of each detected person. A total of 13.3h of RGB, depth, and markerless BPE data are present in the dataset, corresponding to over 1,400,000 frames obtained from a calibrated network with 5 RGB-D cameras. The inertial suits, on the other hand, allowed to record 3h of inertial motion capture data, corresponding to a total of over 600,000 frames recorded by each of the 17 IMUs used by every suit.

Human activity recognition and clinical biomechanics are challenging problems in physical telerehabilitation medicine. However, most publicly available datasets on human body movements cannot be used to study both problems in an out-of-the-lab movement acquisition setting. The objective of the VIDIMU dataset is to pave the way towards affordable patient tracking solutions for remote daily life activities recognition and kinematic analysis. The VIDIMU dataset includes 54 healthy young adults that were recorded on video and 16 of them were simultaneously recorded using custom IMUs. For each subject, 13 activities were registered using a low-resolution video camera and five Inertial Measurement Units (IMUs). Inertial sensors were placed in the lower or the upper limbs of the subject, respectively for activities that involve movement with the lower or the upper body. Video recordings were postprocessed using the state-of-the-art pose estimator BodyTrack (similar to OpenPose, and included in NVIDIA Maxine-AR-SDK) to provide a sequence of 3D joint positions for each movement. Raw IMU recordings were post-processed to compute joint angles by inverse kinematics with OpenSim. For recordings including simultaneous acquisition of video and IMU data types, these signals were used for data file synchronization. Collected data can be further used in applications related to human activity recognition and biomechanics related experiments in simulated home-like settings.